Remote Code Execution

Remote code execution in Google Chrome for iOS prior to version 149.0.7827.53 allows a remote attacker to execute arbitrary code by enticing a user to visit a crafted HTML page. Chromium rates the underlying use-after-free as Critical severity, though SSVC currently shows no observed exploitation and no public exploit identified at time of analysis. The CVSS 8.8 rating reflects high impact across confidentiality, integrity, and availability, tempered by a required user interaction (visiting the malicious page).

Remote code execution in Google Chrome's Ozone display server layer affects all desktop versions prior to 149.0.7827.53, where a use-after-free memory corruption flaw can be triggered by a crafted HTML page. Chromium-internal severity is rated Critical and the CVSS 3.1 score of 8.8 reflects high impact across confidentiality, integrity, and availability, though exploitation requires user interaction (visiting an attacker-controlled page). No public exploit has been identified at time of analysis, and CISA SSVC currently lists Exploitation as 'none'.

Remote code execution in Google Chrome versions prior to 149.0.7827.53 stems from a use-after-free flaw in the Chromoting (Chrome Remote Desktop) component, which Google rated Critical internally. A remote attacker can deliver malicious network traffic to a user with an active Chromoting session and execute arbitrary code in the browser context, though user interaction is required per the CVSS vector. No public exploit identified at time of analysis, and EPSS probability is very low (0.04%).

Remote code execution in Google Chrome's Cast Streaming component (versions prior to 149.0.7827.53) allows an attacker on the same local network segment to execute arbitrary code by sending malicious network traffic to a vulnerable browser. The flaw is rated Critical by the Chromium project and carries a CVSS 3.1 score of 8.8, with no public exploit identified at time of analysis.

Remote code execution in Google Chrome on macOS prior to 149.0.7827.53 stems from a use-after-free flaw in the Chromoting (Chrome Remote Desktop) component, allowing remote attackers to execute arbitrary code by delivering malicious network traffic. Google's Chromium team rates the underlying defect as Critical severity, and no public exploit has been identified at time of analysis, though the bug class historically attracts in-the-wild exploitation against browser users.

Remote code execution in Google Chrome for iOS versions prior to 149.0.7827.53 allows a remote attacker to execute arbitrary code by luring a victim to a crafted HTML page that triggers a use-after-free condition. The flaw is rated Critical by Chromium and carries a CVSS 8.8 score, and while no public exploit is identified at time of analysis, the user-interaction-only barrier (visiting a page) makes drive-by exploitation a realistic concern for unpatched iOS Chrome users.

Remote code execution in Google Chrome versions prior to 149.0.7827.53 stems from a use-after-free flaw in the Network component, allowing a remote attacker to execute arbitrary code within the renderer process when a user visits a crafted HTML page. Google rated this issue Critical at the Chromium level, and a vendor patch is available; no public exploit identified at time of analysis.

Command injection in the stata-mcp (MCP-for-Stata) server versions before 1.17.3 lets attackers execute arbitrary Stata and OS commands by supplying a malicious `log_file_name` to the `stata_do` MCP tool or `stata-mcp tool do` CLI. Because the value is interpolated unsanitized into a Stata `log using` command, an attacker can break out of the quoted string and invoke `shell`, `python`, or `erase`, achieving remote code execution and arbitrary file write while completely bypassing the GuardValidator, which only inspects do-file content. There is no public exploit identified at time of analysis, though the advisory includes a working proof-of-concept payload; EPSS is low at 0.63% (71st percentile).

Out-of-bounds write/read in Zephyr RTOS (versions ≤ 4.3) affects the TLS socket connect path when the TLS session cache is enabled, where tls_session_store() and tls_session_restore() memcpy a caller-supplied socket address into a fixed-size 24-byte stack buffer using an unvalidated, caller-controlled addrlen. Because struct net_sockaddr is opaque, an application can pass an oversized addrlen (e.g. 128 bytes), corrupting adjacent memory and causing a crash/denial of service, with potential for arbitrary code execution. Publicly available exploit code exists per the SSVC 'poc' status, but EPSS is very low (0.06%, 18th percentile) and it is not on CISA KEV.

Local privilege escalation in Seagull Software BarTender 2021 R1 through 12.0.1 allows any low-privileged user on the host to gain SYSTEM execution by sending a crafted BinaryFormatter payload to a localhost-bound .NET Remoting endpoint. Publicly available exploit code exists (a YSoSerial.NET-based PoC is published as a GitHub gist), and the issue carries a CVSS 4.0 base score of 8.5 with high confidentiality, integrity, and availability impact. No CISA KEV listing is present, so exploitation is opportunistic rather than confirmed in-the-wild.

Unauthenticated remote code execution in Seagull Software BarTender 2010, 2016 (<=R9), and 2019 (<=R10) is reachable over TCP/7375 through the BtSystem.Service.exe .NET Remoting endpoint, which runs as NT AUTHORITY\SYSTEM. The exposed singleton uses BinaryServerFormatterSinkProvider with TypeFilterLevel=Full, allowing attackers to abuse deserialization-style object unmarshalling to read/write arbitrary files, coerce NTLMv2 authentication via UNC paths, or pivot to full code execution. No public exploit identified at time of analysis, but the VulnCheck advisory documents the attack primitives in detail.

Arbitrary root code execution in libinput before 1.30.4 and 1.31.x before 1.31.3 arises because the libinput-device-group helper emits the device 'phys' (physical path) property without escaping it. Since this output is consumed by udev - which runs as root - an attacker who controls a device's phys string can inject additional udev properties and have commands executed with root privileges. Publicly available exploit code exists (SSVC exploitation: PoC) but EPSS is very low (0.02%, 6th percentile) and it is not on CISA KEV, indicating no confirmed in-the-wild exploitation.

Remote code execution in Tautulli versions prior to 2.17.1 allows attackers to achieve unauthenticated RCE on fresh installations (pre-setup wizard) by abusing the newsletter custom template directory feature to load a malicious Mako template from an attacker-controlled SMB share. On completed installations the same chain remains exploitable by any authenticated admin. Publicly available exploit code exists per SSVC, and the SSVC framework rates this as automatable with total technical impact, though no CISA KEV listing has been confirmed.

Unauthenticated OS command injection in the API of Progress ADC products - LoadMaster, ECS Connections Manager, Object Scale Connection Manager, and MOVEit WAF - allows remote attackers to execute arbitrary commands on the appliance by supplying unsanitized input to multiple command endpoints. The flaw carries a CVSS of 9.8 (pre-auth, network-reachable) and publicly available exploit code exists (a GitHub PoC and watchTowr Labs write-up), though EPSS remains low at 0.30% and CISA SSVC currently rates exploitation status as 'none'. Fixed builds (7.2.63.2 and 7.2.54.18) are available from Progress.

Authenticated remote command injection in GX Earth ONT (Optical Network Terminal) devices allows attackers with valid web management credentials to execute arbitrary OS commands as root via multiple diagnostic functions in the device's web interface. The flaw, reported by CERT-In and tracked as CIVN-2026-0288, carries a CVSS 4.0 score of 8.7 (High) reflecting low attack complexity and high confidentiality, integrity, and availability impact. No public exploit identified at time of analysis, and the vulnerability is not listed in CISA KEV.

Command injection in the Supermicro AS-2115HS-TNR BMC's SMTP service allows an authenticated administrator to inject crafted characters into the SMTP configuration, causing the underlying system to execute unintended OS commands during process invocation. Successful exploitation can yield arbitrary code execution on the baseboard management controller, denial of service, or persistent compromise of the management plane. No public exploit identified at time of analysis, and the issue is not currently listed in CISA KEV.

Privilege escalation and arbitrary Kubernetes resource creation in Jupyter Enterprise Gateway allows attackers to inject YAML into kernel pod manifests via untrusted KERNEL_* environment variables submitted to the /api/kernels endpoint. Because these values are interpolated into a Jinja2 pod template without YAML-aware escaping, an attacker can clobber the pod securityContext to run kernels as root or use multi-document YAML boundaries (--- / ...) to spin up additional privileged pods, ultimately enabling container escape and full cluster compromise. Publicly available exploit code exists (detailed PoC in the vendor advisory), though EPSS is low (0.06%) and it is not listed in CISA KEV.

Remote code execution in Jupyter Enterprise Gateway (fixed in v3.3.0) is possible because attacker-supplied KERNEL_XXX environment variables (e.g. KERNEL_POD_NAME, KERNEL_WORKING_DIR) sent to the /api/kernels endpoint are rendered through Jinja2 when building the Kubernetes pod manifest, enabling Server-Side Template Injection. Any client able to reach the gateway API can inject Jinja2 expressions to execute Python and OS commands inside the Enterprise Gateway pod, then steal its Kubernetes service account token to read secrets and schedule privileged/hostPath pods, escalating to full cluster compromise. Publicly available exploit code exists (working curl PoCs in the advisory); the flaw is not listed in CISA KEV and EPSS is low at 0.86%.

Privilege bypass in Jupyter Enterprise Gateway versions 2.0.0rc1 through 3.2.x allows remote unauthenticated attackers to launch Jupyter kernels as root (UID/GID 0) by appending whitespace to the KERNEL_UID or KERNEL_GID values, bypassing the EG_PROHIBITED_UIDS/GIDS protection. The flaw chains with Kubernetes hostPath volume mounts to enable container escape and worker-node compromise, with publicly available exploit code (PoC) documented in the GHSA advisory; no public exploitation identified at time of analysis.

Server-side request forgery and potential remote code execution in Docling versions 2.82.0 through 2.90.x affect deployments that explicitly enable the Playwright-based HTML rendering backend. When `render_page=True` is set, untrusted HTML documents can execute arbitrary JavaScript in the rendering browser context and reach internal network services, enabling SSRF, data exfiltration, or code execution in the rendering environment. No public exploit identified at time of analysis, and the issue is not in CISA KEV, but a vendor-confirmed fix exists in 2.91.0.

Path traversal (Zip Slip) in IBM's Docling document processing library before v2.91.0 allows arbitrary file write when the EasyOCR model download function extracts ZIP archives without validating member paths. An attacker who can intercept or substitute the model download source (via MITM, DNS spoofing, or upstream supply-chain compromise) can drop files anywhere the process can write, leading to RCE or persistence. No public exploit identified at time of analysis and the vulnerability is not on the CISA KEV list.

Remote code execution in OpenStack Mistral through version 22.0.0 allows authenticated attackers with low privileges to execute arbitrary code via exposed API endpoints, leading to exfiltration of service credentials and full compromise of the workflow service. The CVSS 9.9 score reflects scope change (S:C) meaning impact extends beyond Mistral itself into other OpenStack components whose credentials it holds. There is no public exploit identified at time of analysis, but the vulnerability is tagged as Authentication Bypass and RCE, and disclosure occurred via oss-security with vendor coordination.

Origin validation bypass in Jupyter Server 1.12.0 through 2.17.0 lets remote attackers defeat CORS origin checks whenever the optional allow_origin_pat configuration is enabled. Because origin matching uses re.match() (anchored only at the start of the string), an attacker-controlled lookalike domain such as trusted.example.com.evil.com satisfies a pattern meant to allow only trusted.example.com, exposing CORS-protected responses, WebSocket kernels, referer checks, and login redirects. Publicly available exploit code exists via the huntr report, but EPSS is very low (0.02%) and SSVC rates exploitation as POC-only and not automatable, so there is no evidence of widespread active exploitation.

Remote code execution in Hugging Face Transformers 5.2.0 allows a malicious model repository to bypass the user's explicit trust_remote_code=False safeguard when loading a LightGlue model via AutoModel.from_pretrained(). The LightGlueConfig deserializes the trust_remote_code flag from the untrusted config.json and propagates the attacker-controlled value into a nested AutoConfig.from_pretrained() call, enabling execution of arbitrary attacker-supplied Python during model initialization. Rated CVSS 9.6 (AV:N/AC:L/PR:N/UI:R) with publicly available exploit code exists via the Huntr disclosure, though EPSS is currently 0.07% (22th percentile) and the CVE is not on CISA KEV.

Server-side environment variable disclosure in MLflow versions prior to 3.11.0 allows attackers to exfiltrate sensitive credentials from the MLflow AI Gateway by abusing the `$ENV_VAR` resolution feature in gateway secret configuration. By registering or modifying a gateway route where `api_key` references an environment variable like `$AWS_SECRET_ACCESS_KEY` and pointing `api_base` at an attacker-controlled endpoint, the resolved secret is transmitted in upstream provider authentication headers. Publicly available exploit code exists via the huntr.com bounty disclosure, though EPSS remains low at 0.28% (51st percentile) and the issue is not in CISA KEV.

Local privilege escalation and arbitrary code execution in Wassimulator CactusViewer v2.3.0 (Windows) occurs through DLL hijacking, where the application loads a malicious DLL from a writable directory instead of the legitimate library. Exploitation requires user interaction to launch the application after an attacker plants the crafted DLL, and no public exploit identified at time of analysis. EPSS rates exploitation probability at just 0.02% (5th percentile), reflecting the niche user base of this open-source image viewer.

Remote code execution in BrowserStack Runner through version 0.9.5 allows network-adjacent unauthenticated attackers to execute arbitrary code on the host system by sending crafted JSON to the /_log HTTP handler. The flaw stems from unsafe use of vm.runInNewContext() combined with eval(), and a known sandbox-escape technique via util.format and this.constructor.constructor enables full host compromise. No public exploit identified at time of analysis, but the technique is well-documented and the CVSS 8.8 score reflects high impact across confidentiality, integrity, and availability.

Arbitrary code execution in the aiohttp Python framework (versions prior to 3.14.0) arises when CookieJar.load() deserializes an attacker-controlled file, classed as CWE-502 unsafe deserialization. An attacker who can plant or substitute the persisted cookie file and induce the application to load it gains code execution in the host process. There is no public exploit identified at time of analysis, EPSS is very low (0.06%), and CISA SSVC rates exploitation as none - consistent with the library's own note that the function is normally used on the user's own trusted data.

Remote code execution in React Router 7.0.0 through 7.14.1 affects applications running in Framework Mode by chaining an application-level prototype pollution flaw with router internals to achieve unauthenticated RCE on the server. Applications using Declarative Mode (BrowserRouter) or Data Mode (createBrowserRouter/RouterProvider) are unaffected. No public exploit identified at time of analysis; CVSS 8.1 reflects high impact tempered by high attack complexity due to the prerequisite prototype pollution gadget.

Remote code execution in the Content Visibility for Divi Builder WordPress plugin (versions ≤ 4.02) allows authenticated attackers with Contributor-level access or higher to execute arbitrary PHP on the server by abusing the 'cvdb_content_visibility_check' parameter of the 'et_pb_text' shortcode. The flaw is a CWE-94 code injection issue disclosed by Wordfence and carries a CVSS 3.1 score of 8.8. No public exploit identified at time of analysis, and the vulnerability is not currently listed in CISA KEV.

Local code execution in NVIDIA NVTabular allows a low-privileged attacker to abuse insecure deserialization of untrusted data, potentially leading to arbitrary code execution, data tampering, and information disclosure on the host running the library. The flaw carries a CVSS 7.8 (High) rating with confidentiality, integrity, and availability all marked High, and currently no public exploit identified at time of analysis. NVTabular is a tabular feature-engineering library used in recommender-system pipelines, so the practical blast radius is data-science workstations and ML training nodes.

Local code execution in NVIDIA NVTabular allows an authenticated low-privileged user to abuse improper deserialization of untrusted data to run arbitrary code, tamper with data, and disclose sensitive information. The CVSS 3.1 base score is 7.8 (AV:L/AC:L/PR:L/UI:N) reflecting a local attack vector with low complexity and low privileges; no public exploit identified at time of analysis and the issue is not on the CISA KEV list.

Unauthenticated remote code execution in Spacelabs Healthcare Sentinel (versions 10.5.x and 11.x prior to 11.6.0) allows network attackers to drop ASPX webshells into the IIS wwwroot via a legacy .NET Remoting HTTP channel on TCP/8989. The flaw stems from missing authentication (CWE-306) on a deprecated Microsoft remoting endpoint, and while a vendor patch is available, no public exploit identified at time of analysis. Exploitation requires that port 8989 has been deliberately exposed to the network, since it is not reachable in default Sentinel deployments.

Arbitrary command execution in Amazon Kiro IDE versions prior to 0.11 allows remote attackers to plant malicious task definitions (e.g., .vscode/tasks.json) via the IDE's file write tool, which then auto-execute when a developer opens the affected workspace folder. The CVSS 4.0 score of 8.6 reflects network-reachable abuse with high confidentiality, integrity, and availability impact contingent on a single user action (folder open). No public exploit identified at time of analysis, and the issue is not listed in CISA KEV.

Remote code execution in OpenMed before 1.5.2 allows unauthenticated attackers to execute arbitrary Python code on the OpenMed service by abusing broad substring matching in the PII privacy-filter model dispatcher to load attacker-controlled Hugging Face repositories with trust_remote_code=True. The flaw, reported by VulnCheck and tracked as CWE-94 (Code Injection), carries a CVSS 4.0 score of 9.3 and a vendor-released patch is available; no public exploit identified at time of analysis.

Remote code execution in Wirtualna Uczelnia (versions up to wu#2016.437.295#0#20260327_105545) allows unauthenticated network attackers to execute arbitrary commands via Server-Side Template Injection in the redirectToUrl endpoint's redirectUrlParameter. The CVSS 4.0 base score of 9.3 reflects no authentication, no user interaction, and high impact across confidentiality, integrity, and availability; no public exploit identified at time of analysis, and the issue is not listed in CISA KEV. Disclosure originated from CERT-PL, indicating a vetted advisory channel for this Polish academic management product.

Authenticated remote code execution in Vivotek FD8136 IP cameras running firmware FD8136-VVTK-0300a allows attackers with valid admin-interface credentials to overflow a buffer in the /cgi-bin/dido/setdo.cgi endpoint and execute arbitrary code as root. No public exploit identified at time of analysis, though a vulnerability research repository on GitHub describes the issue, and EPSS estimates exploitation probability at 0.05% (17th percentile), suggesting low broad-scale exploitation interest despite the high CVSS 8.8 score.

Remote code execution in Vivotek FD8136 IP cameras running firmware FD8136-VVTK-0300a allows authenticated attackers to gain root-level command execution by triggering a buffer overflow in the /cgi-bin/admin/eventtask.cgi admin endpoint. Publicly available exploit code exists in a GitHub research repository, though EPSS scoring (0.09%, 25th percentile) suggests low observed exploitation activity and the CVE is not listed in CISA KEV.

Remote code execution in VIVOTEK FD8136 network camera (firmware VVTK-0300a) is possible via a stack-based buffer overflow in the set_getparam.cgi component, reachable over the network without authentication. CVSS 7.3 reflects partial CIA impact, but the presence of public vulnerability-research artifacts on GitHub indicates publicly available exploit code exists, while EPSS remains low at 0.05% (17th percentile) and the issue is not currently listed in CISA KEV.

Proximity-based remote code execution in Google Android (versions 14, 15, 16, and 16-qpr2) is possible via a heap buffer overflow in multiple functions of sdp_discovery.cc, the Bluetooth Service Discovery Protocol component. An adjacent attacker within Bluetooth range can trigger memory corruption without any user interaction, with no public exploit identified at time of analysis. The flaw resides in the native Bluetooth stack and could yield code execution at the privilege level of the Bluetooth process.

Local privilege escalation in Google Android (versions 14, 15, 16, and 16-qpr2) stems from an integer overflow (CWE-190) that can be triggered without user interaction to achieve code execution. With CVSS 8.4 and SSVC technical impact rated 'total,' a local attacker on the device can elevate privileges across security boundaries without additional execution rights. No public exploit identified at time of analysis and SSVC reports exploitation status as 'none.'

Remote code execution in IBM WebSphere Application Server 9.0 and 8.5 allows authenticated attackers to abuse unsafe Java deserialization in the SAML Web Single Sign-On component to run arbitrary code via a crafted HTTP request combined with a gadget chain. The flaw carries a CVSS 8.5 with scope change, and while no public exploit has been identified at time of analysis, deserialization gadget chains for WebSphere are historically well-researched. IBM has released a patch via support advisory node/7274733.

Remote code execution in IBM WebSphere Application Server 9.0 and 8.5 arises from unsafe deserialization of untrusted data processed by JAX-WS endpoints that use WS-Security. Unauthenticated remote attackers who can reach a SOAP/JAX-WS endpoint may craft malicious serialized payloads to execute arbitrary code in the WebSphere server context. No public exploit identified at time of analysis, but the high CVSS (9.0) and scope-changed impact mean any exposed JAX-WS service is a meaningful target.

Remote code execution in IBM WebSphere Application Server 9.0 and 8.5 allows network-based attackers to bypass security controls and execute arbitrary code on the application server. The flaw carries a CVSS 9.0 critical rating with a scope-changed impact (S:C) and requires no authentication, though high attack complexity (AC:H) suggests specific timing or environmental conditions must be met. No public exploit identified at time of analysis, and the issue is not currently listed in CISA KEV.

Remote code execution in IBM i Access Client Solutions (ACS) versions 1.1.5.0 through 1.1.9.12 allows authenticated attackers to execute arbitrary code on systems where ACS is configured to listen for requests from IBM i Navigator. The flaw, rated CVSS 8.8 with high impact across confidentiality, integrity, and availability, requires low privileges and no user interaction, making it a serious threat in enterprise IBM i environments. There is no public exploit identified at time of analysis and the CVE is not currently in CISA KEV.

Stack-based buffer overflow in rrdcached (the caching daemon for rrdtool) allows a local attacker with socket access to crash the daemon or potentially execute arbitrary code by sending an oversized CREATE request. The flaw is tracked under CWE-121 with a CVSS 3.1 base score of 7.8 (AV:L/AC:L/PR:L), reported by Red Hat against RHEL 6 through 10, and there is no public exploit identified at time of analysis.

Remote code execution in AMD's AI Tensor Engine for ROCm (AITER) through version 0.1.14 allows unauthenticated network attackers to run arbitrary code on every inference worker in a distributed cluster by sending a malicious pickle payload to the ZMQ SUB socket consumed by MessageQueue.recv() in shm_broadcast.py. The vulnerability stems from unauthenticated, unvalidated pickle deserialization with no HMAC or format checks; no public exploit identified at time of analysis, but VulnCheck has published an advisory and AMD has merged an upstream fix.

Secret exfiltration in CloudPirates Open Source Helm Charts (prior to commit fcf9302) allows unauthenticated attackers to steal repository secrets, including Docker Hub credentials and tokens, by submitting a malicious fork pull request that triggers the pull-request.yaml GitHub Actions workflow in a privileged context. No public exploit is identified at time of analysis, but the attack pattern is a well-known pwn-request misconfiguration that requires no maintainer approval, and the CVSS score of 10.0 (Critical) with scope change reflects compromise of CI/CD secrets that extend beyond the workflow boundary.

Credential exposure in CloudPirates Open Source Helm Charts (prior to commit fcf9302) allows attackers to steal a Personal Access Token and SSH signing key by submitting a malicious pull request that the vulnerable GitHub Actions workflow checks out and executes with privileged secrets. The repository's generate-schema.yaml workflow used unsafe pull_request_target patterns combined with checkout of fork-controlled code, enabling code injection into a privileged CI context. No public exploit identified at time of analysis, but the attack pattern is well-documented for this class of GitHub Actions misconfiguration.

Local code execution in Poppler's Splash rendering backend allows attackers to compromise applications that open attacker-supplied PDFs by triggering an integer overflow in tilingPatternFill that produces an undersized heap allocation and a subsequent out-of-bounds write. The flaw affects Poppler as shipped across Red Hat Enterprise Linux 6 through 10 and Red Hat Hardened Images, with impact including arbitrary code execution, information disclosure, or denial of service in the rendering process. No public exploit identified at time of analysis, and the CVSS 7.8 vector requires user interaction to open a malicious PDF.

Remote code execution in Disig Web Signer versions 2.0.3 through 2.5.3 allows network-adjacent attackers to execute arbitrary code on systems running this Slovak electronic signing client when a user performs a passive interaction (e.g., visiting a malicious page or processing a crafted signing request). The flaw was reported by Slovakia's National Security Authority (incident@nbu.gov.sk) and carries a CVSS 4.0 score of 9.4 (Critical) with both vulnerable and subsequent system impact rated High, indicating a likely scope-changing condition. There is no public exploit identified at time of analysis and the vulnerability is not currently listed in the CISA KEV catalog.

Remote code execution in Poly Voice products on Linux is possible through a stack-based buffer overflow reachable when administrators enable Interactive Connectivity Establishment (ICE). Unauthenticated network attackers can trigger the flaw without user interaction, and no public exploit has been identified at time of analysis. HP rates the issue at CVSS 4.0 9.2 (Critical), driven by network reachability and full confidentiality, integrity, and availability impact.

Reflected cross-site scripting in Vitest browser mode (@vitest/browser) allows attackers to execute arbitrary JavaScript in the Vitest server origin by luring a developer to a crafted /__vitest_test__/ URL with a malicious otelCarrier query parameter. Because the same page embeds VITEST_API_TOKEN used to authenticate the Vitest WebSocket API, the XSS chains into full Node-side remote code execution by writing to vite.config.ts and triggering a config reload. Publicly available exploit code exists in the vendor's GHSA-2h32-95rg-cppp advisory, though no public exploit identified at time of analysis in CISA KEV.

Unauthenticated remote code execution in Dassault Systèmes Teamwork Cloud (No Magic Release 2022x-2026x) and Magic Collaboration Studio (CATIA Magic Release 2022x-2026x) arises from unsafe deserialization of attacker-controlled data. The CVSS 9.8 vector indicates a network-reachable attack with no privileges or user interaction, yielding full confidentiality, integrity, and availability impact, though no public exploit identified at time of analysis and EPSS data was not provided.

Remote code execution in Apache ActiveMQ Broker, ActiveMQ All, and ActiveMQ (versions before 5.19.7 and 6.0.0 before 6.2.6) allows authenticated attackers to bypass the CVE-2026-34197 fix using non-parenthesized discovery wrappers such as `masterslave:vm://...` and `static:vm://...`, which incorrectly pass validation and trigger the VM transport's brokerConfig parameter to load a remote Spring XML application context. The flaw abuses the Jolokia JMX-HTTP bridge at /api/jolokia/ to invoke BrokerService.addNetworkConnector/addConnector MBean operations, resulting in arbitrary code execution on the broker JVM. EPSS is low at 0.06% (19th percentile) and no public exploit identified at time of analysis, but the patch bypass nature and prior in-the-wild interest in ActiveMQ RCE chains warrant urgent patching.

Authenticated remote code execution in Apache ActiveMQ Classic (versions before 5.19.7 and 6.0.0 through 6.2.5) is achievable via the Jolokia JMX-HTTP bridge exposed at /api/jolokia/ on the web console. An authenticated attacker can invoke BrokerService.addNetworkConnector with a crafted masterslave:// discovery URI that loads a Spring XML application context, instantiating attacker-controlled singleton beans (e.g., Runtime.exec()) on the broker JVM. No public exploit identified at time of analysis and EPSS is low (0.06%), but the vendor-released patches and CVSS 8.1 score reflect a significant risk to message brokers exposing the web console.

Heap buffer overflow in the MediaTek WLAN access point driver allows adjacent-network attackers with low-privilege user execution to corrupt memory and achieve remote code execution without user interaction. The flaw affects multiple MediaTek Wi-Fi chipsets commonly embedded in routers and access points (MT7615, MT7915, MT7916, MT7981, MT7986, MT7990, MT7992, MT7993, MT6890). No public exploit identified at time of analysis, and the issue is not listed in CISA KEV.

Arbitrary code execution in ESA AnomalyMatch before 1.3.1 allows local attackers with low privileges to run code under the application's process by planting a malicious PyTorch checkpoint into a session directory, which is loaded via torch.load() with weights_only=False. No public exploit is identified at time of analysis, but the upstream fix (PR #9) and a third-party advisory (imlabs.info) confirm the unsafe-deserialization root cause and the migration to safetensors.

Remote code execution in the Spectra Gutenberg Blocks WordPress plugin (versions up to and including 2.19.25) allows authenticated users with Contributor-level access or higher to execute arbitrary PHP on the server by abusing the plugin's block rendering logic. The flaw stems from the plugin trusting attacker-controlled block attributes to register a fake uagb/-prefixed block type with an arbitrary render_callback, which is then invoked via call_user_func() when a second block of the same type is rendered in the same request. No public exploit identified at time of analysis, but the low privilege bar (Contributor is commonly granted to untrusted guest authors) makes this a high-priority issue for sites with open registration.

Pre-authentication full account takeover in praisonai-platform (pip package, versions <= 0.1.2) allows remote unauthenticated attackers to forge JWTs for any user, including workspace owners and admins. The JWT signing key silently falls back to the hardcoded literal 'dev-secret-change-me' from the public GitHub source because the production-mode guard only triggers when PLATFORM_ENV is explicitly set to a non-default value, which default deployments do not do. Publicly available exploit code exists in the GHSA advisory itself, though no public exploit campaign or CISA KEV listing has been reported at time of analysis.

Remote code execution in PraisonAI's first-party A2A server example (pip package <= 4.6.39) allows unauthenticated network attackers to execute arbitrary Python on the server process by sending a JSON-RPC message/send request to /a2a that drives the LLM-backed agent into invoking an eval()-based calculate tool. The chain combines three first-party defaults - no auth_token, 0.0.0.0 bind, and an eval()-backed example tool - and was confirmed end-to-end against a real Gemini model with a canary marker file write. No public exploit identified at time of analysis beyond the proof-of-concept in the advisory itself, and the issue is not in CISA KEV.

Remote code execution in PraisonAI praisonaiagents <=1.6.39 and PraisonAI <=4.6.39 allows authenticated attackers to fully escape the execute_code() subprocess sandbox by leveraging print.__self__ to reach the real builtins module and reconstructing __import__ at runtime. The flaw defeats prior patches for CVE-2026-39888, CVE-2026-34938, and CVE-2026-40158, enabling arbitrary OS command execution on the host wherever agent input can be influenced via prompt injection or direct code submission. Publicly available exploit code exists in the GitHub Security Advisory (GHSA-4mr5-g6f9-cfrh), and EPSS/KEV signals are not yet published for this newly disclosed novel bypass.

Remote code execution in PraisonAI versions 2.0.0 through 4.6.39 allows attackers to execute arbitrary Python code via two unguarded spec.loader.exec_module call sites in agents_generator.py. The flaw is a sibling of CVE-2026-44334: the v4.6.32 chokepoint refactor added a PRAISONAI_ALLOW_LOCAL_TOOLS env-var gate to tool_override.py but missed the load_tools_from_module and load_tools_from_module_class sinks, which load arbitrary module paths from YAML agent configuration without validation. Publicly available exploit code exists (working PoC published with the advisory) and a fixed release is available in 4.6.40.

Remote code execution in ouroboros-ai (pip package) allows arbitrary command execution when a developer runs any Ouroboros command inside a cloned malicious repository. The tool unconditionally loaded `.env` from the current working directory with full trust, enabling an attacker-controlled `OUROBOROS_CLI_PATH` or similar variable to redirect subprocess calls to a malicious script. No public exploit or active exploitation (CISA KEV) has been confirmed; the fix is available in version 0.39.0 and the EPSS score of 0.56% (69th percentile) indicates moderate exploitation likelihood relative to the broader CVE population.

Heap buffer overflow write in FreeRDP's server-side clipboard (cliprdr) channel allows a malicious RDP client to crash server processes and potentially achieve remote code execution against FreeRDP versions prior to 3.26.0. The flaw is triggered by a malformed CB_CLIP_CAPS PDU with an undersized capabilitySetLength field, corrupting heap memory after authentication. No public exploit identified at time of analysis, but the CVSS 8.8 rating and heap corruption nature make this a high-priority patch for any RDP server deployment using FreeRDP.

Heap buffer overflow write in FreeRDP client versions prior to 3.26.0 allows a malicious RDP server to corrupt client memory and potentially achieve code execution when the victim connects with RDPGFX enabled. The flaw resides in gdi_CacheToSurface, where validation uses a clamped destination rectangle while the actual copy uses unclamped cacheEntry width/height values, enabling a large out-of-bounds heap write. No public exploit identified at time of analysis, and the issue is not listed in CISA KEV.

Code execution in JetBrains IntelliJ IDEA prior to 2026.1 is achievable through server-side template injection (SSTI) in the built-in Copyright plugin, which lets an attacker embed executable template expressions into copyright profile configurations. Because the CVSS vector is local with required user interaction (AV:L/UI:R), exploitation hinges on a developer opening or importing a malicious project/profile that triggers the injected template, running with the victim's privileges. There is no public exploit identified at time of analysis and EPSS is 0.00%, consistent with the SSVC 'exploitation: none' assessment.

Remote code execution in JetBrains TeamCity before 2026.1 allows an authenticated user with project/build configuration rights to run arbitrary code on the CI server by tampering with Perforce (P4) connection settings. The flaw stems from unsafe handling of version-control connection parameters (CWE-88, argument injection), and while it carries a high CVSS of 8.8 with total technical impact, there is no public exploit identified at time of analysis and its EPSS probability is 0.00%.

Sandbox escape leading to remote code execution in vm2 NodeVM versions 3.11.3 and earlier allows attackers running untrusted JavaScript inside the sandbox to break out via two missing entries in the dangerous-builtin denylist: `process` (whose `getBuiltinModule()` reloads any core module, including `child_process`) and `inspector/promises` (whose `Session().post('Runtime.evaluate', ...)` evaluates code in the host realm). The flaw is exploitable only when the embedder allows `process`, `inspector/promises`, or wildcard `*` in `require.builtin`. Publicly available exploit code exists (PoC published with the advisory), and a patch is available in vm2 3.11.4; no public exploit identified at time of analysis as actively used in the wild.

Sandbox escape in vm2 (npm package, versions <= 3.11.3) allows arbitrary code execution in the Node.js host process when untrusted code is run on runtimes exposing WebAssembly JSPI (Node 24 with --experimental-wasm-jspi, or Node 26+ by default). A working PoC demonstrates that a JSPI-backed Promise reaches host-realm Promise.prototype.finally without bridge interposition, letting attacker-controlled species logic walk a rejection object's constructor chain to host process and execute arbitrary commands. No public exploit identified as actively used in the wild, but a complete weaponized PoC is published in the GHSA advisory.

Sandbox escape in vm2 (npm package, versions <= 3.11.3) allows full remote code execution by bypassing the GHSA-8hg8-63c5-gwmx (CVE-2023-37903) patch through omission of the require option when nesting:true is set. Publicly available exploit code exists in the GHSA advisory itself, and with a CVSS of 10.0 (AV:N/AC:L/PR:N/UI:N/S:C) any application passing untrusted code to a NodeVM configured this way is trivially compromised on the host.

Sandbox escape in vm2 (npm) versions ≤ 3.11.3 allows untrusted code running inside the VM to write arbitrary properties - including dangerous Symbol-keyed properties like `nodejs.util.promisify.custom` - onto host-realm objects via a Proxy `set`-trap receiver-handling flaw. Because any inheriting child object (`Object.create(hostObj)`) becomes a write channel into the host, attackers can hijack host control flow when the host later dispatches through the polluted slot, escalating from sandbox integrity loss to host-realm code execution. Publicly documented reproduction exists in the GHSA advisory; no public exploit identified at time of analysis and not on the CISA KEV list.

Remote code execution in the vm2 Node.js sandbox library (versions <= 3.11.3) allows sandboxed JavaScript to escape isolation and execute arbitrary commands on the host. The flaw stems from a missing resetPromiseSpecies call in the localPromise swallow tail, letting a sandbox subclass hijack Promise species resolution and capture a raw host-realm RangeError that exposes the host Function constructor. Publicly available exploit code exists (POC published in the advisory), though no public exploit identified at time of analysis as actively exploited in the wild; given vm2's heavy use as an untrusted-code sandbox in SaaS, CI, and serverless platforms, this is a top-priority issue.

Sandbox escape in the vm2 Node.js sandbox library (versions <= 3.11.3) allows untrusted JavaScript executed inside a vm2 VM to reach the host realm and achieve arbitrary code execution. By chaining Buffer.call.call indirection with __lookupGetter__/__lookupSetter__ to obtain host Object.prototype.__proto__ accessors and using a WebAssembly.compileStreaming rejection to surface a host TypeError, an attacker severs a host prototype chain, causing the bridge's proto-walk to return the raw host error unwrapped - yielding e.constructor.constructor as host Function and full RCE. Publicly available exploit code exists (advisory ships a working PoC); CVSS is 10.0 (AV:N/AC:L/PR:N/UI:N/S:C/C:H/I:H/A:H).

Authenticated path traversal in Dokploy v0.26.5 and earlier (CWE-22) enables arbitrary file write during application deployment, escalating to remote code execution when the affected instance uses the remote server deployment feature. With a CVSS 9.9 score reflecting scope change and full CIA impact, any user with deployment privileges can drop cron jobs onto remote hosts to fully compromise them, bypassing container isolation. No public exploit identified at time of analysis, though the vendor's own GHSA-66v7-g3fh-47h3 advisory characterizes the chain as critical.

Credential theft and response hijacking is possible in Axios (npm) versions 1.0.0-1.15.1 and 0.19.0-0.30.x when a co-resident prototype pollution vulnerability allows an attacker to place a JavaScript function on Object.prototype.transformResponse before Axios processes a request. The Axios mergeConfig() function reads config values through normal property access, allowing the prototype-inherited function to displace the legitimate response transform; when executed by transformData(), the function runs with the full request config as this, exposing auth credentials, URLs, and headers, while also replacing the response returned to application code. A publicly available exploit (PoC) exists per the GitHub Security Advisory GHSA-3g43-6gmg-66jw and SSVC assessment, but no confirmed active exploitation (CISA KEV) has been recorded; EPSS is 0.05% (15th percentile), consistent with a chained, non-trivial attack.

Remote code execution in manga-image-translator's share.py API server allows unauthenticated attackers to achieve full container compromise via unsafe pickle deserialization. The /execute/{method_name} and /simple_execute/{method_name} endpoints deserialize attacker-controlled request bodies with pickle.loads(), enabling arbitrary code execution in the server process. Publicly available exploit code exists, but the EPSS score (0.36%) indicates low current exploitation probability.

Cross-origin JavaScript injection in Home Assistant Companion apps for Android (before 2026.4.4) and iOS (before 2026.4.1) allows a malicious iframe loaded inside the app's WebView to execute arbitrary JavaScript in the Home Assistant frontend's origin and steal the signed-in user's access token. The flaw stems from exposing the native JavaScript bridge (window.externalApp on Android, webkit.messageHandlers.getExternalAuth on iOS) to all frames combined with unsanitized callback identifier interpolation. No public exploit identified at time of analysis, and the issue is not in CISA KEV.

Wallet balance forgery in WWBN AVideo 29.0 and earlier allows any authenticated user to credit their own wallet with arbitrary funds by abusing an unfinished AuthorizeNet payment handler. The plugin/AuthorizeNet/processPayment.json.php endpoint hardcodes payment success and calls YPTWallet::addBalance() using an attacker-supplied amount with no validation of the underlying Authorize.Net transaction. No public exploit has been identified at time of analysis, but the flaw is trivially reproducible from a logged-in account.

Arbitrary code execution in Roslyn CodeLens MCP Server versions 0.0.9 through 1.17.0 allows attackers to run code in the MCP server process by tricking a victim into opening a malicious .NET solution. The get_diagnostics tool automatically loads and executes any DiagnosticAnalyzer DLL referenced by a .csproj without allowlist, signature validation, or user prompt, and includeAnalyzers defaults to true. No public exploit identified at time of analysis, but the attack pattern (malicious project file + attacker-controlled DLL) is well understood and trivial to weaponize.

Arbitrary file upload in Interinfo DreamMaker allows authenticated high-privilege remote attackers to upload web shell backdoors and achieve arbitrary code execution on the underlying server. The flaw carries a CVSS 4.0 score of 8.6 and was reported via TWCERT, with no public exploit identified at time of analysis. Despite the high privilege requirement, the network-reachable vector and full CIA impact on the host make this a meaningful post-authentication compromise primitive.

Unauthenticated arbitrary file upload in Interinfo DreamMaker allows remote attackers to upload web shell backdoors and execute arbitrary code on the server, scoring CVSS 4.0 9.3 (Critical). No public exploit identified at time of analysis, but the combination of no authentication, low attack complexity, and complete CIA impact makes this a high-priority issue. TWCERT (Taiwan's national CERT) is the reporting source, indicating regional advisory coordination.

Remote code execution in Mautic 7 allows authenticated users with the campaign:imports:create permission to write arbitrary PHP files outside the intended extraction directory via a malicious ZIP uploaded through the campaign import feature. The flaw is a path traversal in the ZIP extraction routine that can be leveraged to overwrite cache or configuration files, yielding code execution as the web server user. No public exploit identified at time of analysis, and the CVSS 9.9 score reflects the combination of low attack complexity, scope change, and full CIA impact.

Server-Side Template Injection in Mautic's theme engine allows authenticated users with theme creation or upload permissions to execute arbitrary code on the host. Because the platform renders uploaded Twig templates without a sandbox or strict function allowlist, an attacker who already holds theme-management rights can pivot from administrative content control to full server compromise, including reads of restricted system files and configuration. No public exploit identified at time of analysis, and the issue has not been listed in CISA KEV.

Remote unauthenticated command injection in the Acer Predator Connect W6x router enables root-level code execution via crafted MQTT messages, scoring a maximum 10.0 CVSS v4.0 across all impact metrics including subsequent system impact. The flaw was disclosed by Acer and tracked in the ENISA EUVD as EUVD-2026-33269, but no public exploit identified at time of analysis. Affects firmware W6x_GBL_2.00.000005 and earlier, with no special authentication or user interaction required.

Local privilege escalation in ASUS System Control Interface allows an authenticated low-privileged user to gain SYSTEM-level code execution by issuing a crafted RPC call that bypasses the component's validation logic. The flaw stems from incorrect permission assignment (CWE-732) on a critical resource exposed over RPC, and no public exploit identified at time of analysis. CVSS 4.0 scores it 7.3 with high attack complexity, indicating exploitation is non-trivial but yields full host compromise.

Remote code execution in Falco Solutions PHPageBuilder v0.31.0 is possible through an unrestricted file upload flaw in the pagemanager/pagebuilder module, enabling unauthenticated remote attackers to upload arbitrary executable files (typically PHP webshells) and run code on the underlying web server. Publicly available exploit code exists in a dedicated GitHub repository, though EPSS scoring (0.06%, 18th percentile) and SSVC indicate no observed in-the-wild exploitation despite the attack being fully automatable.

Command injection in Dulwich (pure-Python Git implementation) versions >= 0.24.0 and < 1.2.5 allows remote attackers to execute arbitrary OS commands when a victim merges an attacker-controlled branch and has a custom merge driver configured that references the %P placeholder. The ProcessMergeDriver passes attacker-controlled file paths from the git tree into subprocess.run with shell=True, so a path like 'x; touch /tmp/pwned #' is interpreted as a shell metacharacter sequence. Publicly available exploit code exists (working POC in the GitHub Security Advisory GHSA-9277-mp7x-85jf); no public exploit in CISA KEV at time of analysis.

Arbitrary file write leading to remote code execution in Dulwich (pure-Python Git implementation) versions >= 0.10.0 and < 1.2.5 allows attackers controlling a Git repository to plant files inside a Windows victim's .git directory - most notably .git/hooks/pre-commit.exe - which executes on the next commit. The flaw stems from the NTFS path-element validator accepting Windows-hostile bytes (\, :, and git~<n> 8.3 short-name aliases) plus broken handling of core.protectNTFS/core.protectHFS configuration. No public exploit identified at time of analysis, and the issue is not listed in CISA KEV, but the technique is closely modeled on the well-documented Git CVE-2019-1353/1354 class.

Type confusion in the V8 JavaScript engine of Google Chrome before 148.0.7778.216 enables sandboxed arbitrary code execution when a user is convinced to install a malicious browser extension that delivers a crafted payload. The flaw carries a CVSS 7.5 (AV:N/AC:H/PR:N/UI:R) and is reported by Google's Chrome team, with no public exploit identified at time of analysis and a very low EPSS of 0.02%. Impact remains constrained to the renderer sandbox, but full confidentiality, integrity, and availability loss inside that boundary is possible.